- cookies
            - client
            - state
            - url_for
            - client_state
            - application_state
            - receive
            - send
            - accept
            - receive_text
            - receive_bytes
            - receive_json
            - iter_text
            - iter_bytes
            - iter_json
            - send_text
            - send_bytes

            - cookies
            - client
            - state
            - url_for
            - client_state
            - application_state
            - receive
            - send
            - accept
            - receive_text
            - receive_bytes
            - receive_json
            - iter_text
            - iter_bytes
            - iter_json
            - send_text
            - send_bytes

Wenn ein eingehender Request nicht korrekt validiert wird, wird eine „400“-Response gesendet.

## `GZipMiddleware`

Verarbeitet GZip-Responses für alle Requests, die `"gzip"` im `Accept-Encoding`-Header enthalten.

Diese Middleware verarbeitet sowohl Standard- als auch Streaming-Responses.

```Python hl_lines="2  6"
{!../../../docs_src/advanced_middleware/tutorial003.py!}
```

        session: Annotated[str | None, Cookie()] = None,
        item_id: str,
    ):
        if session is None:
            raise WebSocketException(code=status.WS_1008_POLICY_VIOLATION)
        await websocket.accept()
        while True:
            data = await websocket.receive_text()
            await websocket.send_text(f"Session cookie is: {session}")

    },
)
async def create_item(request: Request):
    raw_body = await request.body()
    try:
        data = yaml.safe_load(raw_body)
    except yaml.YAMLError:
        raise HTTPException(status_code=422, detail="Invalid YAML")
    try:
        item = Item.model_validate(data)
    except ValidationError as e:
        raise HTTPException(status_code=422, detail=e.errors(include_url=False))

        username: str = payload.get("sub")
        if username is None:
            raise credentials_exception
        token_data = TokenData(username=username)
    except JWTError:
        raise credentials_exception
    user = get_user(fake_users_db, username=token_data.username)
    if user is None:
        raise credentials_exception
    return user

        username: str = payload.get("sub")
        if username is None:
            raise credentials_exception
        token_data = TokenData(username=username)
    except JWTError:
        raise credentials_exception
    user = get_user(fake_users_db, username=token_data.username)
    if user is None:
        raise credentials_exception
    return user

    但本例仍然可行，而且本例展示了如何与内部组件进行交互。

同样也可以在异常处理器中访问请求体。

此时要做的只是处理 `try`/`except` 中的请求：

```Python hl_lines="13  15"
{!../../../docs_src/custom_request_and_route/tutorial002.py!}
```

发生异常时，`Request` 实例仍在作用域内，因此处理错误时可以读取和使用请求体：

```Python hl_lines="16-18"

                "required": required,
                "model_config": model_config,
                "alias": alias,
            }
        )
    try:
        return ModelField(**kwargs)  # type: ignore[arg-type]
    except (RuntimeError, PydanticSchemaGenerationError):
        raise fastapi.exceptions.FastAPIError(
            "Invalid args for response field! Hint: "
            f"check that {type_} is a valid Pydantic field type. "

```

支持以下参数：

* `allowed_hosts` - 允许的域名（主机名）列表。`*.example.com` 等通配符域名可以匹配子域名，或使用 `allowed_hosts=["*"]` 允许任意主机名，或省略中间件。

如果传入的请求没有通过验证，则发送 `400` 响应。

## `GZipMiddleware`

处理 `Accept-Encoding` 请求头中包含 `gzip` 请求的 GZip 响应。

中间件会处理标准响应与流响应。

```Python hl_lines="2  6"
{!../../../docs_src/advanced_middleware/tutorial003.py!}
```

支持以下参数：

* `minimum_size` - 小于最小字节的响应不使用 GZip。 默认值是 `500`。